org.jaitools.numeric
Class SampleStats

java.lang.Object
  org.jaitools.numeric.SampleStats

public class SampleStats
extends Object

A collection of static methods to calculate summary statistics for a sample of double-valued data. This class is used by both Jiffle and the KernelStats operator.

Since:

1.0

Version:

$Id$

Author:

Michael Bedward, Daniele Romagnoli, GeoSolutions S.A.S.

Constructor Summary

SampleStats()

Method Summary

static double	max(Double[] values, boolean ignoreNaN) Return the maximum of the given values.
static double	mean(Double[] values, boolean ignoreNaN) Return the mean of the given values.
static double	median(Double[] values, boolean ignoreNaN) Calculates the median of the given values.
static double	min(Double[] values, boolean ignoreNaN) Calculates the minimum of the given values.
static double	mode(Double[] values, boolean ignoreNaN) Calculates the empirical mode (highest frequency value) of the given values.
static double	range(Double[] values, boolean ignoreNaN) Calculates the range (max - min) of a set of values.
static double	sdev(Double[] values, boolean ignoreNaN) Calculates sample standard deviation.
static double	sum(Double[] values, boolean ignoreNaN) Calculates the sum of the values.
static double	variance(Double[] values, boolean ignoreNaN) Calculates sample variance using the running sample algorithm of Welford (1962) described by Knuth in The Art of Computer Programming (3rd ed) Vol.2 p.232

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SampleStats

public SampleStats()

Method Detail

max

public static double max(Double[] values,
                         boolean ignoreNaN)

Return the maximum of the given values.

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

max value or Double.NaN if the sample is empty

mean

public static double mean(Double[] values,
                          boolean ignoreNaN)

Return the mean of the given values.

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

mean value or Double.NaN if the sample is empty

min

public static double min(Double[] values,
                         boolean ignoreNaN)

Calculates the minimum of the given values.

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

min value or Double.NaN if the sample is empty

median

public static double median(Double[] values,
                            boolean ignoreNaN)

Calculates the median of the given values. For a sample with an odd number of elements the median is the mid-point value of the sorted sample. For an even number of elements it is the mean of the two values on either side of the mid-point.

Parameters:

values - sample values (need not be pre-sorted)

ignoreNaN - specifies whether to ignore NaN values

Returns:

median value or Double.NaN if the sample is empty

mode

public static double mode(Double[] values,
                          boolean ignoreNaN)

Calculates the empirical mode (highest frequency value) of the given values. Double.NaN values are ignored. If more than one data value occurs with maximum frequency the following tie-break rules are used:

• for an odd number of tied values, return their median
• for an even number of tied values, return the value below the mid-point of the sorted list of tied values

This ensures that the calculated mode occurs in the sample data. Whether or not the mode is meaningful for the sample is up to the user !

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

calculated mode or Double.NaN if the sample is empty

range

public static double range(Double[] values,
                           boolean ignoreNaN)

Calculates the range (max - min) of a set of values.

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

the range or Double.NaN if the set is empty

variance

public static double variance(Double[] values,
                              boolean ignoreNaN)

Calculates sample variance using the running sample algorithm of Welford (1962) described by Knuth in The Art of Computer Programming (3rd ed) Vol.2 p.232

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

sample variance

sdev

public static double sdev(Double[] values,
                          boolean ignoreNaN)

Calculates sample standard deviation. This is a convenience method that calls variance(java.lang.Double[], boolean) and returns the square-root of the result

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

sample standard deviation as a double

sum

public static double sum(Double[] values,
                         boolean ignoreNaN)

Calculates the sum of the values.

Parameters:

values - sample values

ignoreNaN - specifies whether to ignore NaN values

Returns:

sum of the values